Intelligent robotic navigation for elderly assistance

• Main Author: Ng, Joshua Jeremiah
• Other Authors: Smitha Kavallur Pisharath Gopi
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Computer science and engineering::Hardware::Control structures and microprogramming; Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision; Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
• Online Access: https://hdl.handle.net/10356/147914
• Institution: Nanyang Technological University

With the advancement in robotic technologies such as object detection, computer vision and automation, the use of robots in the healthcare sector is increasingly prevalent. One such example is the SESTO HealthGUARD, an autonomous mobile disinfectant robot, which can self-navigate to desired locations and eliminate 99.9% of bacteria, germs and viruses [1]. This robotic technology can also be brought into the common household, targeting the people who are vulnerable, the elderly. The report presents the final year project to develop and functionally test robotic navigation using Turtlebot3 for elderly assistance. The final goal is to create an autonomous robot that can track the elderly’s movement, detect when he/she has fallen, and notify the caretaker. The project consists of 4 phases. The first phase is to test the Turtlebot3 robotic system, which includes its mobility, sensors, and mapping and navigation capabilities. The second phase is computer vision, getting the robot to correctly identify the elderly, as well as detecting when the elderly has fallen. The third phase is tracking, the robot must keep the elderly in its field of view and search for him/her otherwise. The final phase is adding a notification system, allowing the robot to notify the caretaker of the emergency.